Model: GELU Model: 4 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 699 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.8435. Min Act: -0.1700

Data Index: 1012436 (C4 (Web Text))

Max Activating Token Index: 20

Text #1

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.7184. Min Act: -0.1700

Data Index: 890721 (C4 (Web Text))

Max Activating Token Index: 500

Text #2

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.5931. Min Act: -0.1700

Data Index: 515964 (C4 (Web Text))

Max Activating Token Index: 399

Text #3

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.5931. Min Act: -0.1700

Data Index: 884531 (C4 (Web Text))

Max Activating Token Index: 414

Text #4

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.6088. Min Act: -0.1700

Data Index: 509219 (C4 (Web Text))

Max Activating Token Index: 283

Text #5

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.5618. Min Act: -0.1700

Data Index: 1308508 (C4 (Web Text))

Max Activating Token Index: 959

Text #6

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.5305. Min Act: -0.1700

Data Index: 1032667 (C4 (Web Text))

Max Activating Token Index: 68

Text #7

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4992. Min Act: -0.1700

Data Index: 307911 (C4 (Web Text))

Max Activating Token Index: 565

Text #8

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.5305. Min Act: -0.1700

Data Index: 125263 (C4 (Web Text))

Max Activating Token Index: 843

Text #9

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4678. Min Act: -0.1700

Data Index: 1184448 (C4 (Web Text))

Max Activating Token Index: 445

Text #10

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4678. Min Act: -0.1700

Data Index: 905407 (C4 (Web Text))

Max Activating Token Index: 248

Text #11

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4678. Min Act: -0.1700

Data Index: 789717 (C4 (Web Text))

Max Activating Token Index: 966

Text #12

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4521. Min Act: -0.1700

Data Index: 1112974 (C4 (Web Text))

Max Activating Token Index: 882

Text #13

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4365. Min Act: -0.1700

Data Index: 392964 (C4 (Web Text))

Max Activating Token Index: 10

Text #14

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4678. Min Act: -0.1700

Data Index: 832861 (C4 (Web Text))

Max Activating Token Index: 1002

Text #15

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4208. Min Act: -0.1700

Data Index: 1252209 (C4 (Web Text))

Max Activating Token Index: 262

Text #16

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4208. Min Act: -0.1700

Data Index: 1229886 (C4 (Web Text))

Max Activating Token Index: 460

Text #17

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.3894. Min Act: -0.1700

Data Index: 1200916 (C4 (Web Text))

Max Activating Token Index: 959

Text #18

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.4051. Min Act: -0.1700

Data Index: 1195613 (C4 (Web Text))

Max Activating Token Index: 460

Text #19

Max Range: 3.8435. Min Range: -3.8435

Max Act: 3.3894. Min Act: -0.1700

Data Index: 1342317 (C4 (Web Text))

Max Activating Token Index: 22